In last weekend's Barron's, research firm GaveKal outlined the bull case for permanently higher profit margins and thus stock prices (summary here). Fund manager John Hussman rebutted that argument in his weekly essay on the U.S. market. This is critical reading for long term investors:
GaveKal Research recently published a piece that was highly critical of the idea that profit margins revert to the mean over time. Though their bit wasn't directed at me personally, my eyebrows did pop up at their remark that the profit margin argument “annoys us to no end, for it smacks of either lazy thinking and lack of work, or worse, downright intellectual dishonesty.”
They continue, “Indeed, take a look at the graph below.” (Click on charts to enlarge.)
“Now, looking at this chart, can our reader identify any "mean-reverting" tendency? Neither can we.”
Well guys, take a look at the graph below (which looks pretty much the same as the one above):
Now, looking at this chart, can our reader identify any "mean-reverting" tendency? Neither can we. Problem is, this graph is the S&P 500 price/peak earnings ratio from 1950 to 2000. Had you looked at that graph, the relentless uptrend of the prior decade might have convinced a casual observer that valuations don't revert to the mean over time, but instead can simply drift higher indefinitely.
Yet notice that prior to the late-90's blowoff, valuations were well-contained below 20 times peak earnings. Neither high valuations nor low valuations were permanent. The same has been true throughout market history. Clearly, adding a small out-of-range segment to a normally mean-reverting chart can make it look (at least temporarily) as if the mean reversion doesn't exist.
Oh, and in the two years following the end of this particular chart, the price/peak earnings multiple on the S&P 500 fell in half.
Now go back to the first chart, which presents gross after-tax cash flows as a percentage of GDP. You can see the same pattern. Note that until about 1995, this chart also had a clear mean-reverting tendency. Cash flows as a share of GDP were actually very well contained by an upper bound of 12%. Over the past decade, however, the cash flow share has broken free of its prior 12% upper bound, pushing to a record high over 15% presently.
According to GaveKal, a major reason for the “sustained rise in US after-tax margins” is that corporate taxes have dropped sharply since the 1960's and 1970's. While their claim about taxes is correct – corporate taxes as a share of pre-tax income have dropped from nearly 50% during the 70's to about 33% today – virtually all of that decline in corporate tax burdens occurred prior to 1990. Indeed, the tax share of corporate income was about 35% in 1989, as well as 1995, and again, is about 33% today. So while taxes have come down since the 60's, the recent surge in margins can hardly be attributed to that cause. Nor can it be attributed to a reduction in interest burdens, which experienced the majority of their decline well before the recent surge in margins.
Most importantly, GaveKal suggests, the rise in margins is owed to a purported shift to a “knowledge based economy,” where evidently, we all just hang around and consume each others' knowledge. Let's see. I finished my Christmas shopping on Friday. I got in my car, which had a full tank of gas, and having filled my tires earlier in the week at a gas station (where I bought some gum), I drove to the mall, bought some toys, clothes, music, and other gifts, had a cup of coffee, stopped at the food court for a bite to eat, then drove home, checked my answering machine, which had a message from the dentist about a checkup, which reminded me I should probably schedule a haircut in a week or two. At that point, I made some tea, took a vitamin, sat down at my desk, and turned on my computer to write this piece. While it's true that there's a good amount of knowledge and information embodied in various products, the fact is that most of the people and businesses I've come into contact with have been in the business of producing tangible goods or functional services. And of course that's what most people in the U.S. create for a living. They produce stuff or they do stuff.
Sure, there are certainly great benefits from knowledge-based technologies, but as information economists have long noted, as soon as information is produced, one person can consume it without reducing others' consumption of it. The marginal cost of extra copies is zero, and market prices tend to move toward marginal cost. So it's possible to earn a profit on information for a while, but in a competitive economy, it's extraordinarily difficult to retain those profits. Instead, you move toward a “free disposal” equilibrium where people use and enjoy the information but there are no excess profits available from that use. Companies can use computers and the internet, but they still have to compete with other companies that do the same thing. Broadly speaking, the notion that some grand shift to a knowledge-based economy will afford everyone with permanently higher profit margins just strikes me as an unfounded platitude.
So why have profit margins expanded in recent years? Very simply, because labor compensation – which is also mean-reverting – has stagnated (though that's started to change in recent months, both in the U.S. and internationally, thanks to relatively low unemployment).
To see how closely profits and labor compensation are related, I've reprinted a chart that appeared in these comments a few months ago. The blue line (right scale) depicts U.S. corporate profits as a percentage of nominal GDP. The violet line (left scale, smoothed) depicts U.S. personal disposable income as a percentage of nominal GDP, using an inverted scale – a rising line means a falling disposable income share. Notice that increasing corporate profits as a share of GDP generally come at the expense of wage earners' share, and vice versa. The recent upleg in corporate profits since 2003 reflects a corresponding drop in personal income as a share of GDP (a rising violet line) from 75% to 72% of GDP.
The recent rise in profit margins is not the permanent dynamic of some brave new world, but rather a quite cyclical dynamic pushed to a temporary extreme.
Moreover, our views about the cyclical nature of profit margins are based on the very dynamics of capitalism. When we say “margins will return to the mean,” what we are saying is the same thing that economist Joseph Schumpeter said half a century ago: the emergence and elimination of excess profits through the competitive mechanism is not simply a passing phenomenon, but is the essential fact that drives the economy forward. If high profit margins are permanent, then the free enterprise system is a failure.
(Note to the guys at GaveKal – I don't intend this piece as an offense. You write some good, thought-provoking stuff. It's just when you diminish concerns about profit margins by lobbing out phrases like “lazy thinking, lack of work, or worse, downright intellectual dishonesty,” well Tucky, them's fightin' words in these here parts.)